Intelligent maintenance management model using reinforcement algorithm applied to process industry

Abstract

newline Industry 4.0 has grabbed impetus and set a tone of conversation among stakeholders not only in wealthy nations but also in developing economies over the past five years. The most prominent reason for the elaboration of Industry is its implicit impact on society, the economy and manufacturing. In todayand#8223;s many, manufacturing industry antedate a significant impact of Industry 4.0 on their operations and maintenance. Thus, maintenance has gained significance as a support function for ensuring equipment availability, quality products, on-time deliveries and plant safety based on quantitative and qualitative literature review and findings of empirical studies, a conceptual framework is developed for the adoption of Industry has in manufacturing organizations operating in Rubber industry. A crucial feature of intelligent Maintenance perpetration, grounded on literature survey and empirical observation, is the performing maintenance job redesign which purportedly draws upon creativity and culminates in a fortified terrain of growth and provocation. The specific outcome made out of this thesis is the implementation strategies developed by using the Markov decision process, Reinforcement Learning algorithm and Analytic Hierarchy Process for the selected rubber industry running twenty-four hours/day, throughout a year without any obstacles during production by applying the proposed intelligent maintenance management system. The Markov decision process and Reinforcement Learning methods offered comparable results with overall equipment efficiency of 87.16% and 73.84% for short-term maintenance and long term maintenance respectively in the selected module of the rubber industry.